- Email: [email protected]
The effect of tacrolimus on colon anastomotic healing
GASTROENTEROLOGYVol. 114, No. 4
Al178 AGAABSTRACTS • G4817
THE EFFECT OF TACROLIMUS ON COLON ANASTOMOTIC HEALING. M. Sch[iffer, J. V/51ker, B. Proksch, M. Bongartz, H.D. Becker, Tiibingen, Germany. We have shown that the immunosuppressant tacrolimus (TA), used in solid organ transplantation, decreases wound mechanical strength and collagen deposition in dermal wound healing~ Failure of colonic healing has a major impact on surgical morbidity and mortality. Whether tacrolimus affects the outcome of colonic healing, however, is not known. We, therefore, studied the effect of systemic treatment with tacrolimus on standard parameters of colonic wound healing in rats. Groups of eight Sprague Dawley rats underwent laparotomy and single layer 6-0 prolene inverted left colon anastomosis. Beginning at the day of wounding, rats were treated once daily by s.c. injections with 2.0, or 5.0 mg TA/kg. Rats were sacrificed at postoperative day five for determination of colonic bursting pressure (BP) and hydroxyproline content of anastomoses (OHP), an index of reparative collagen deposition. Splenic lymphocytes were tested for proliferative activity (3H-thymidine uptake ). Tacrolimus levels in blood, obtained by cardiac puncture at the day of sacrific e, were measured by ELISA. OHP BP Tacrolimuslevels in blood (ng) (N) (ng/ml) Control 11.7 -+ 1.7 103 -+8 not detected TA2.0 10.7_+.+1.1 82-+15 18.8±3.1" TA5.0 9.5_+0.8 109_+15 43.2-+5.9*# Mean -+SEM, n = 8, *p < 0.01 vs. Co, #p< 0.01 vs. TA 2.0, ANOVA Systemic treatment with tacrolimus was equally well tolerated by all animals. At the time of sacrifice there were no signs of anastomotic leak or abscess formation. Proliferative activity of splenic lymphocytes was significantly decreased by systemic treatment with TA (data not shown), indicating that the tacr01imus doses used were immunosuppressive. Tacrolimus did not affect wound collagen accumulation or bursting strength of colon anastomoses. In contrast, as previously reported, a dose of 2.0 mg TA/kg/day was found to impair dermal healing. Our data show that systemic treatment with tacrolimus at immunosuppressive doses does not affect colonic bursting pressure and wound collagen accumulation, suggesting that differential mechanisms may regulate the outcome of healing in dermal and colonic repair. • G4818
SOMATOSTATIN RECEPTOR SUBTYPES 1 AND 2 COUPLE DIFFERENTIALLY TO ADENYLYLCYCLASES AND TO VOLTAGEGATED CALCIUM CHANNELS IN PANCREATIC RIN 1046-38 CELLS. H. Schertibl 1, D. Roosterman2, G. Glassmeier 1, H. Baumeister 2, E.-O. Riecken l, W. Meyerhofz. IAbt. Innere MedizirdGastroenterologie, Universit~itsklinikum Benjamin Franklin, Freie Universit~tt Berlin, Berlin, 2Abt. Molekulare Genetik, Dentsches Institut ftir Em~ihmngsforschung und Universit~it Potsdam, Potsdam, Germany. Most neuroendocrine gastrointestinal tumors express somatostatin receptors in a high density. This is the basis for both the diagnostic and therapeutic use of somatostatin analogs. The inhibitory effects of somatostatin receptor activation on the hypersecretory activity are mediated by various signal transduction pathways. Rat insulinoma 1046-38 cells represent a model system to study pancreatic neuroendocrine cell function. To characterize the coupling of somatostatin receptor subtypes 1 and 2 (sstj, sst2) to voltagegated Ca2+-channels and adenylylcyclases, we performed RT-PCR, displacement binding and patch-clamp experiments in PIN 1046-38 cells. The mRNAs for sst I and sst2, two of the five somatostatin receptors, were detected by reverse transcription polymerase chain reaction amplification in these cells. Displacement binding analysis suggested that sstI represented the major somatostatin receptor subtype. The ssq selective compound CH-275 did not inhibit adenylyl cyclases in PIN 1046-38 cells, while compounds that activated sst2 did. Performing patch-clamp experiments in PIN 1046-38 cells, CH-275 was found to strongly inhibit voltage-gated Ca2+-channels, while the sst2 specific analog octreotide elicited a less pronounced effect. This suggests that in pancreatic PIN 1046-38 cells the inhibition of voltage-gated Ca2+-channels and thus of Ca2+ dependent hormone release is preferentially mediated by sst r The project was supported by the Deutsche Forschungsgemeinschaft (Sche 326/3-1). • G4819
ENDOGENOUS GLP-l(7.36)AMIDE CONTROLS ENDOCRINE PANCREATIC SECRETION AND ANTRODUODENAL MOTILITY IN HUMAN. J. Schirra. R. Roggel, P. Leicht, U. Wank, R. Arnold, B. G/Ske, M. Katschinski. Dept. of Gastroenterology, University of Marburg, Germany. Exogenous glucagon-like peptide-l(7-36)amide (GLP-1) stimulates insulin secretion, decreases glucagon secretion, and inhibits gastric emptying. The functions of endogenous GLP-1 are unknown. Exendin(9-39)amide (Ex-9) has recently been characterized as a potent, specific and competitive antagonist at the human GLP-1 receptor in vitro and in vivo. Therefore, we used this antagonist to examine the endocrine and motor effects of
endogenous GLP-1 in human in the interdigestive state and with step doses of intraduodenal glucose. Methods: 9 healthy male volunteers in random order underwent 2 series of 2 experiments each. In the first series, an interdigestive period of 30 min was followed by intraduodenal glucose perfusion at 1 kcal/min for 60 min, and at 2.5 kcal/min for another 60 min against an intravenous background of Ex-9 at 300 pmol.kg-Lmin -1 or saline. These glucose loads are below and above the threshold for GLP-1 release, respectively (J Clin Invest 1996; 97:92-103). Antroduodenal motor (perfusion manometry, transmucosal potential difference measurement) and endocrine responses were assessed. To calculate the incretin effect of GLP-1, two additional experiments were performed in each volunteer with intravenous glucose infusion mimicking plasma glucose levels obtained with duodenal glucose perfusion. Results: Mean-+ SEM, *:P<0.05 vs saline, #:P<0.05 vs interdigestive, §:P<0.05 vs duodenal glucose at 1 kcal/min Interdigestive I Duodenal Glucose Perfusion
Blood glucose (mg/dl,meanoverbasal) Insulin ~aU/l,meanoverbasal) [ncretin effect '%) ~31ueagon '.pg/ml,meanoverbasal) 3LP-1
'pmoFl,meanoverbasal) ~ntral motility .~ontractions/10rain) Duodenum :ontraetions/10min)
Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9
0.5 ± 0.8 7.2 ± 2.1* 0.08 ± 0.2 0.16 ± 0.2 -5.6 ± 2.6 12.7 ± 2.4* 0.3 ± 0.4 -0.2 ± 0.2 4.8 ± 1.5 12.1 ± 1.9" 31.0 ± 4.8 54.7 ± 9.2*
1 keal/min 19.4 ± 1.5# 24.9 ± 2.6*# 10.4 ± 1.4# 9.0 ± 1.6# 48.5 ± 8.2 26.0 ± 6.5* -13.2 ± 2.0# 2.6 ± 2.7*# 0.0 ± 0.1 -0.2 ± 0.5 2.3 ± 0.5# 7.8 ± 1A* 24.4 ± 2.2# 42.9 ± 5.9*
2.5 kcal/min 41.1 ± 3.4§ 53.9 ± 4.7*§ 47.9 ± 6.7§ 36.7 ± 6.4*§ 69.5 ± 5.6§ 47.1 ± 5.9*§ -22.3 ± 4.4§ 0.9 ± 3.8*§ 3.2 ± 0.9§ 6.2 ± 1.2"§ 1.5 ± 0.5§ 2.8 ± 0.6*§ 14.2 ± 2.3§ 32.7 ± 4.6*#
With infusion of Ex-9, 11 antral phases III occurred compared to 2 antral phases III with saline. Conclusions: Basal plasma levels of endogenous GLP-1 tonically inhibit glucagon release thus lowering postabsorptive blood glucose. GLP-1 is not released with perfusion of glucose at 1 kcal/min. Nevertheless, basal GLP-I reduces glycemic responses to low duodenal glucose loads by stimulating glycemia-related insulin secretion and inhibiting glucagon secretion. Basal GLP-1 markedly contributes to the incretin effect with low duodenal glucose loads, Duodenal glucose at 2.5 kcal/min stimulates GLP-1 secretion resulting in diminished glycemic excursion as a consequence of enhanced insulin secretion. This implies an increased incretin effect and, more important, a lowered glucagon release. There appears to be an auto-feedback regulation of GLP-1 release in human. Basal GLP-1 is an inhibitory modulator of antroduodenal motility in the interdigestive state and in response to low duodenal glucose loads. Stimulation of GLP-1 release by high duodenal glucose loads reveals a considerable inhibitory effect of GLP-1 on the intestinal phase of postprandial antroduodenal motility. However, the glucose mediated inhibition of antroduodenal motility is also brought about by mechanisms independent of GLP-1, arguably by direct interaction with intestinal glucoreceptors. • G4820
EXENDIN(9-39)AMIDE, BUT NOT EXENDIN(3-39)AMIDE, IS AN ANTAGONIST OF GLP-l(7-36)AMIDE IN HUMAN. J..Schjrra: M. Junck, G. Kraft, H. Schmidt, U. Wank, P. Leicht, B. Gtike, M. Katschinski. Dept. of Gastroenterology, University of Marburg, Germany. Exendin(9-39)amide (Ex-9) is a potent, specific and competitive receptor antagonist of glucagon-like peptide-l(7-36)amide (GLP-1) in human. Recently, exendin(3-39)amide (Ex-3) was suggested as an up to 10-fold more potent antagonist of GLP-1 in rat (J Biol Chem 1997;272:21201-6). GLP-1 increases intracellular cAMP in the pancreatic B-cell and stimulates insulin secretion. Furthermore, exogenous GLP-1 decreases the tone of the proximal stomach (Wank et al., submitted to DDW 1998). To characterize the potential of Ex-3 as a GLP-1 antagonist in human, we compared binding characteristics and intracellular cAMP response in human insulinoma (HUI) ceils in vitro, and insulinotropic as well as motor response in vivo of Ex-3 and Ex-9 with and without GLP- 1. Methods: In vitro studies: HUI cells were assayed for binding characteristics (displacement of 125I-GLP-1) as well as changes in intraeellular cAMP induced by the peptides tested in the absence or presence of different stimulatory doses of GLP-1. In the experiments with GLP-1, concentration of both exendins was 10-6 M. In vivo studies: 2 healthy male volunteers underwent 4 experiments in random order. After a basal period of 30 min, GLP-1 was IV infused at 0.3 pmolokg-l.min -1 mimicking physiological postprandial plasma levels during 30 min in euglycemia, and thereafter for another 60 min with a concomitant infusion of exogenous glucose, yielding a stable hyperglycemia of 8 mmol/l. On different days, saline, Ex-3 at 300 pmol°kg-l.min -l, or Ex-9 at 300 pmolokg-l°min-I were IV infused as background. A fourth day with sole infusion of saline served as control. The motor response of the proximal stomach during euglycemia (electronic barostat, capacity of intragastric bag 1I00 ml, means over basal) and plasma insulin levels during hyperglycemia (absolute means) were assessed.
Al178 AGAABSTRACTS • G4817
THE EFFECT OF TACROLIMUS ON COLON ANASTOMOTIC HEALING. M. Sch[iffer, J. V/51ker, B. Proksch, M. Bongartz, H.D. Becker, Tiibingen, Germany. We have shown that the immunosuppressant tacrolimus (TA), used in solid organ transplantation, decreases wound mechanical strength and collagen deposition in dermal wound healing~ Failure of colonic healing has a major impact on surgical morbidity and mortality. Whether tacrolimus affects the outcome of colonic healing, however, is not known. We, therefore, studied the effect of systemic treatment with tacrolimus on standard parameters of colonic wound healing in rats. Groups of eight Sprague Dawley rats underwent laparotomy and single layer 6-0 prolene inverted left colon anastomosis. Beginning at the day of wounding, rats were treated once daily by s.c. injections with 2.0, or 5.0 mg TA/kg. Rats were sacrificed at postoperative day five for determination of colonic bursting pressure (BP) and hydroxyproline content of anastomoses (OHP), an index of reparative collagen deposition. Splenic lymphocytes were tested for proliferative activity (3H-thymidine uptake ). Tacrolimus levels in blood, obtained by cardiac puncture at the day of sacrific e, were measured by ELISA. OHP BP Tacrolimuslevels in blood (ng) (N) (ng/ml) Control 11.7 -+ 1.7 103 -+8 not detected TA2.0 10.7_+.+1.1 82-+15 18.8±3.1" TA5.0 9.5_+0.8 109_+15 43.2-+5.9*# Mean -+SEM, n = 8, *p < 0.01 vs. Co, #p< 0.01 vs. TA 2.0, ANOVA Systemic treatment with tacrolimus was equally well tolerated by all animals. At the time of sacrifice there were no signs of anastomotic leak or abscess formation. Proliferative activity of splenic lymphocytes was significantly decreased by systemic treatment with TA (data not shown), indicating that the tacr01imus doses used were immunosuppressive. Tacrolimus did not affect wound collagen accumulation or bursting strength of colon anastomoses. In contrast, as previously reported, a dose of 2.0 mg TA/kg/day was found to impair dermal healing. Our data show that systemic treatment with tacrolimus at immunosuppressive doses does not affect colonic bursting pressure and wound collagen accumulation, suggesting that differential mechanisms may regulate the outcome of healing in dermal and colonic repair. • G4818
SOMATOSTATIN RECEPTOR SUBTYPES 1 AND 2 COUPLE DIFFERENTIALLY TO ADENYLYLCYCLASES AND TO VOLTAGEGATED CALCIUM CHANNELS IN PANCREATIC RIN 1046-38 CELLS. H. Schertibl 1, D. Roosterman2, G. Glassmeier 1, H. Baumeister 2, E.-O. Riecken l, W. Meyerhofz. IAbt. Innere MedizirdGastroenterologie, Universit~itsklinikum Benjamin Franklin, Freie Universit~tt Berlin, Berlin, 2Abt. Molekulare Genetik, Dentsches Institut ftir Em~ihmngsforschung und Universit~it Potsdam, Potsdam, Germany. Most neuroendocrine gastrointestinal tumors express somatostatin receptors in a high density. This is the basis for both the diagnostic and therapeutic use of somatostatin analogs. The inhibitory effects of somatostatin receptor activation on the hypersecretory activity are mediated by various signal transduction pathways. Rat insulinoma 1046-38 cells represent a model system to study pancreatic neuroendocrine cell function. To characterize the coupling of somatostatin receptor subtypes 1 and 2 (sstj, sst2) to voltagegated Ca2+-channels and adenylylcyclases, we performed RT-PCR, displacement binding and patch-clamp experiments in PIN 1046-38 cells. The mRNAs for sst I and sst2, two of the five somatostatin receptors, were detected by reverse transcription polymerase chain reaction amplification in these cells. Displacement binding analysis suggested that sstI represented the major somatostatin receptor subtype. The ssq selective compound CH-275 did not inhibit adenylyl cyclases in PIN 1046-38 cells, while compounds that activated sst2 did. Performing patch-clamp experiments in PIN 1046-38 cells, CH-275 was found to strongly inhibit voltage-gated Ca2+-channels, while the sst2 specific analog octreotide elicited a less pronounced effect. This suggests that in pancreatic PIN 1046-38 cells the inhibition of voltage-gated Ca2+-channels and thus of Ca2+ dependent hormone release is preferentially mediated by sst r The project was supported by the Deutsche Forschungsgemeinschaft (Sche 326/3-1). • G4819
ENDOGENOUS GLP-l(7.36)AMIDE CONTROLS ENDOCRINE PANCREATIC SECRETION AND ANTRODUODENAL MOTILITY IN HUMAN. J. Schirra. R. Roggel, P. Leicht, U. Wank, R. Arnold, B. G/Ske, M. Katschinski. Dept. of Gastroenterology, University of Marburg, Germany. Exogenous glucagon-like peptide-l(7-36)amide (GLP-1) stimulates insulin secretion, decreases glucagon secretion, and inhibits gastric emptying. The functions of endogenous GLP-1 are unknown. Exendin(9-39)amide (Ex-9) has recently been characterized as a potent, specific and competitive antagonist at the human GLP-1 receptor in vitro and in vivo. Therefore, we used this antagonist to examine the endocrine and motor effects of
endogenous GLP-1 in human in the interdigestive state and with step doses of intraduodenal glucose. Methods: 9 healthy male volunteers in random order underwent 2 series of 2 experiments each. In the first series, an interdigestive period of 30 min was followed by intraduodenal glucose perfusion at 1 kcal/min for 60 min, and at 2.5 kcal/min for another 60 min against an intravenous background of Ex-9 at 300 pmol.kg-Lmin -1 or saline. These glucose loads are below and above the threshold for GLP-1 release, respectively (J Clin Invest 1996; 97:92-103). Antroduodenal motor (perfusion manometry, transmucosal potential difference measurement) and endocrine responses were assessed. To calculate the incretin effect of GLP-1, two additional experiments were performed in each volunteer with intravenous glucose infusion mimicking plasma glucose levels obtained with duodenal glucose perfusion. Results: Mean-+ SEM, *:P<0.05 vs saline, #:P<0.05 vs interdigestive, §:P<0.05 vs duodenal glucose at 1 kcal/min Interdigestive I Duodenal Glucose Perfusion
Blood glucose (mg/dl,meanoverbasal) Insulin ~aU/l,meanoverbasal) [ncretin effect '%) ~31ueagon '.pg/ml,meanoverbasal) 3LP-1
'pmoFl,meanoverbasal) ~ntral motility .~ontractions/10rain) Duodenum :ontraetions/10min)
Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9 Saline Ex-9
0.5 ± 0.8 7.2 ± 2.1* 0.08 ± 0.2 0.16 ± 0.2 -5.6 ± 2.6 12.7 ± 2.4* 0.3 ± 0.4 -0.2 ± 0.2 4.8 ± 1.5 12.1 ± 1.9" 31.0 ± 4.8 54.7 ± 9.2*
1 keal/min 19.4 ± 1.5# 24.9 ± 2.6*# 10.4 ± 1.4# 9.0 ± 1.6# 48.5 ± 8.2 26.0 ± 6.5* -13.2 ± 2.0# 2.6 ± 2.7*# 0.0 ± 0.1 -0.2 ± 0.5 2.3 ± 0.5# 7.8 ± 1A* 24.4 ± 2.2# 42.9 ± 5.9*
2.5 kcal/min 41.1 ± 3.4§ 53.9 ± 4.7*§ 47.9 ± 6.7§ 36.7 ± 6.4*§ 69.5 ± 5.6§ 47.1 ± 5.9*§ -22.3 ± 4.4§ 0.9 ± 3.8*§ 3.2 ± 0.9§ 6.2 ± 1.2"§ 1.5 ± 0.5§ 2.8 ± 0.6*§ 14.2 ± 2.3§ 32.7 ± 4.6*#
With infusion of Ex-9, 11 antral phases III occurred compared to 2 antral phases III with saline. Conclusions: Basal plasma levels of endogenous GLP-1 tonically inhibit glucagon release thus lowering postabsorptive blood glucose. GLP-1 is not released with perfusion of glucose at 1 kcal/min. Nevertheless, basal GLP-I reduces glycemic responses to low duodenal glucose loads by stimulating glycemia-related insulin secretion and inhibiting glucagon secretion. Basal GLP-1 markedly contributes to the incretin effect with low duodenal glucose loads, Duodenal glucose at 2.5 kcal/min stimulates GLP-1 secretion resulting in diminished glycemic excursion as a consequence of enhanced insulin secretion. This implies an increased incretin effect and, more important, a lowered glucagon release. There appears to be an auto-feedback regulation of GLP-1 release in human. Basal GLP-1 is an inhibitory modulator of antroduodenal motility in the interdigestive state and in response to low duodenal glucose loads. Stimulation of GLP-1 release by high duodenal glucose loads reveals a considerable inhibitory effect of GLP-1 on the intestinal phase of postprandial antroduodenal motility. However, the glucose mediated inhibition of antroduodenal motility is also brought about by mechanisms independent of GLP-1, arguably by direct interaction with intestinal glucoreceptors. • G4820
EXENDIN(9-39)AMIDE, BUT NOT EXENDIN(3-39)AMIDE, IS AN ANTAGONIST OF GLP-l(7-36)AMIDE IN HUMAN. J..Schjrra: M. Junck, G. Kraft, H. Schmidt, U. Wank, P. Leicht, B. Gtike, M. Katschinski. Dept. of Gastroenterology, University of Marburg, Germany. Exendin(9-39)amide (Ex-9) is a potent, specific and competitive receptor antagonist of glucagon-like peptide-l(7-36)amide (GLP-1) in human. Recently, exendin(3-39)amide (Ex-3) was suggested as an up to 10-fold more potent antagonist of GLP-1 in rat (J Biol Chem 1997;272:21201-6). GLP-1 increases intracellular cAMP in the pancreatic B-cell and stimulates insulin secretion. Furthermore, exogenous GLP-1 decreases the tone of the proximal stomach (Wank et al., submitted to DDW 1998). To characterize the potential of Ex-3 as a GLP-1 antagonist in human, we compared binding characteristics and intracellular cAMP response in human insulinoma (HUI) ceils in vitro, and insulinotropic as well as motor response in vivo of Ex-3 and Ex-9 with and without GLP- 1. Methods: In vitro studies: HUI cells were assayed for binding characteristics (displacement of 125I-GLP-1) as well as changes in intraeellular cAMP induced by the peptides tested in the absence or presence of different stimulatory doses of GLP-1. In the experiments with GLP-1, concentration of both exendins was 10-6 M. In vivo studies: 2 healthy male volunteers underwent 4 experiments in random order. After a basal period of 30 min, GLP-1 was IV infused at 0.3 pmolokg-l.min -1 mimicking physiological postprandial plasma levels during 30 min in euglycemia, and thereafter for another 60 min with a concomitant infusion of exogenous glucose, yielding a stable hyperglycemia of 8 mmol/l. On different days, saline, Ex-3 at 300 pmol°kg-l.min -l, or Ex-9 at 300 pmolokg-l°min-I were IV infused as background. A fourth day with sole infusion of saline served as control. The motor response of the proximal stomach during euglycemia (electronic barostat, capacity of intragastric bag 1I00 ml, means over basal) and plasma insulin levels during hyperglycemia (absolute means) were assessed.